SXR Technical Design Report - Stanford University

SOFT X--RAY MATERIAL INSTRUMENT ((SXR))
SUMMARY OF THE TECHNICAL DESIGN
DRAFT 3-3-09
This document is derived from the SXR Technical Design Report, July 2008, prepared for the
LCLS by the SXR consortium. Sections 1 to 4 are the same. Section 5 has been revised to
represent the current status of the design. Section 6+ the end station requirements and interfaces
still to be revised.

OUTLINE SXR - Summary of the Technical Design:
Page
1. Introduction 3
2. Scientific Program 4
2.1 Pump-Probe Ultrafast Chemistry 4
2.2 Clusters as new Materials 5
2.3 Magnetic Imaging 6
2.4 X-ray Scattering Spectroscopy on Strongly Correlated Materials 8
2.5 High-Resolution Ultrafast Coherent Imaging 9
3. Science Driven Requirements 11
3.1 Experimental stations 11
3.2 Monochromator 11
3.3 Focusing optics 12
3.4 Diagnostics 12
3.5 Laser systems 12
4. Optical design 14
4.1 Optical layout 14
4.2 Optical elements 14
4.3 Raytracing, energy resolution and efficiency 17
5. Instrument Layout 20
5.1 Instrument Configuration 20
5.2 Single Pulse Shutter 23
5.3 Transmission Sample Station 23
5.4 Monochromator 24
5.5 Zero Order Stop and Grating Apertures 24
5.6 Exit Slit Collimators 24
5.7 Exit Slits 25
5.8 Spectrograph Detector 25
5.9 Photon Stopper 25
5.10 Intensity Monitors 26
5.11 Refocusing Mirrors 26
5.12 Experimental End Station 27
5.13 Optical Pump Laser Transport 28
Appendix: C Descriptions of experimental chambers including detectors from groups that
propose to bring systems.
xx
2

1. INTRODUCTION
The Linac Coherent Light Sources (LCLS) is building the first x-ray Free Electron Laser (FEL) where light is
scheduled to be delivered in August 2009. The facility will operate in the wave length range of 1.5 nm-0.15 nm (800
eV-8 KeV). The proposed soft x-ray imaging and pump-probe x-ray spectroscopy program on materials was
approved by the LCLS Scientific Advisory Committee (SAC) in 2006, and space was allocated in the LCLS near
hall for the accompanying instruments. Due to budget constraints, the materials soft x-ray program did not receive
any instrumentation funding from LUSI. We here propose a plan to move forward with the soft x-ray program by
constructing an instrument that contains a monochromator, refocusing optics and various locations where
experimental endstations can utilize the beam in the soft x-ray energy regime.
In order to fund, design and construct the beam line we have formed a consortium with members from the
Stanford Institute of Material and Energy Sciences (SIMES), the Advanced Light Source (ALS), University of
Hamburg-DESY, Cener for Free Electron Lasers (CFEL) in Hamburg and BESSY in Berlin. The consortium will
serve as a central hub for a broader scientific community with interest in utilizing the unique capabilities of the
LCLS in the soft x-ray regime. The consortium and various collaborators will also bring different endstations and
detectors to the instrument for the experimental program that can also be utilized by the general users.
Since LCLS will be delivering light at longer wave length in its early operation and the completion of several
LUSI instruments is delayed it is essential that instruments exists that can make use of the early LCLS beam. We
therefore plan to move forward in a timely fashion with this effort with the goal to have an instrument ready in late
summer of 2009. In the following we will describe the science, parameters, optical and technical design of the
instrument. We will in the appendix include a short description of the various experimental endstations that will be
brought to the instrument and letters of intent from the broader scientific community demonstrating their interest.
3

2. SCIENTIFIC PROGRAM
2.1 Pump-Probe Ultrafast Chemistry
The ultimate goal in chemistry and physical chemistry is to understand on a fundamental level how bonds break
and reform during chemical reactions. In many cases we arrive at simple pictures of electron motion with respect to
electron pair redistributions or electrostatic interactions along a reaction path. For many systems bonding can be
understood in terms of molecular orbitals and reactivity in dynamical rearrangements of different molecular states.
Such knowledge provides the basis for the understanding of chemical trends and prediction of chemical reactivity
for chemical compounds. Since the excitation and probe steps with conventional optical lasers involve valence
electrons that are delocalized over many atomic centers it is difficult to study complex systems. Unprecedented
insight into chemical reaction dynamics would be gained by probing exactly the atomic site involved. X-ray
spectroscopies can directly access molecular orbital changes associated with or even during chemical reactions. In
particular, accessing core levels in the soft x-ray regime with spectroscopy opens up new prospects to study timeresolved
changes in the electronic structure of complex systems containing the essential elements C, O and N or 3dmetal
atoms. Detailed insight into surface reactions, catalysis, hydrogen-bonded systems and aqueous solutions can
be expected.
Figure 0-1. Schematic illustration of X-ray Spectroscopy. Here, the radiative decay of a core-hole in N 2
adsorbed on a Ni surface by x-ray emission spectroscopy is shown. In addition to elemental sensitivity the
method provides specificity to different chemical sites as shown in this example.
X-ray spectroscopy has the unique ability to provide an atom-specific probe of the electronic structure. In x-ray
emission spectroscopy (XES) the atomic or elemental sensitivity arises from the filling of a core hole by valence
electrons from the same atomic site. In addition, core-level energy shifts (often denoted chemical shifts) connected
with different environments allow for selective probing of chemically non-equivalent atoms (Figue 0-1). The final
state of the x-ray emission process is a valence-hole state similar to the final state in valence band photoemission
with the unique feature that the valence electronic structure is projected onto a specific atom. Notably, selection
rules of XES, and similarly of X-ray photoelectron spectroscopy (XPS), in conjunction with variation of polarization
4

vector of the incident light or angle-resolved detection of electrons allows to access molecular orbital symmetry and
associated bond geometry. In addition, XPS can be uniquely tuned to high surface sensitivity, which is particularly
desirable when studying interfaces, including the aqueous/vacuum interface. Resonant excitation and Auger electron
spectroscopy gives unique access to the electronic structure of atoms and molecules in the gas phase, on surfaces
and in liquids and solids. Some of the projects planned here are described in further detail in the following:
Surface Reactions and Catalysis: The microscopic understanding of heterogeneous catalysis requires a
detailed understanding of the dynamics of elementary processes at surfaces including adsorption, formation of
different intermediates, and desorption. These can be initiated by an ultrashort (optical) laser pulse, and the evolving
product can be uniquely probed with XES and XPS using FEL soft x-ray pulses at a given time delay. Charge and
energy transfer processes, for instance between an adsorbate and the (catalytic) substrate, can thus be studied with
high site-selectivity and high temporal resolution. Likewise, we expect to identify and to characterize chemical
bonding in short-lived reaction intermediates by transient changes of the electronic structure, from which kinetic
models with an unprecedented level of detail can be derived.
Hydrogen Bonding, Radiation and Aqueous Solution Chemistry: Water is the key species for our existence
on earth, and it is involved in nearly all biological, geological, and chemical processes. Knowledge about the
hydrogen-bonded network structure in liquid water is essential for understanding its unusual chemical and physical
properties. Infrared and optical excitations will be used to induce changes in the hydrogen bonding network, and to
initiate reactions of hydrated molecules. Moreover, on a pratical level, radiation damage of concentrated electrolyte
solutions, relevant for fuel storage, involves the anions interacting with ionizing radiation. To give another example,
low-energy electrons created upon laser irradiation may attach to DNA bases in aqueous environment. Dissociative
electron attachment is a major cause of strand breaks in DNA. Hydrogen-bond dynamics, solute-solvent interactions,
and the interplay of electronic and nuclear dynamics during chemical reactions in solution will be accessible with
time-resolved optical pump and x-ray probe spectroscopy. Techniques include XAS, XES, and XPS measurements;
the latter will be performed in conjunction with the liquid microjet technique.
Warm Dense Matter: Warm Dense Matter refers to the region of the density-temperature phase-space
between solids and plasmas, where the standard theories of condensed matter physics and/or plasma statistical
physics are invalid. [x] These states of matter are of broad interest, as similar conditions of high temperatures and
pressures exist in planetary interiors and in shock compressed matter. Further, there is an incomplete understanding
of how materials damage under Free Electron Laser (FEL) irradiation. A femtosecond optical or FEL pulse will
isochorically heat the sample. The x-ray emission will yield information about the occupied density of states. The
objective is to develop a quantitative understanding of the electronic structure of materials at solid densities and
temperatures of several 1000 ˚K.
x. T. Ao, Y. Ping, K. Widmann, D.F. Price, E. Lee, H. Tam, P.T. Springer and A. Ng, Phys. Rev. Lett. 96, 55001
(2006).
2.2 Clusters as new Materials
Materials assembled atom by atom offer the chance to create new materials with controlled but as yet
unprecedented new properties. In the nanometer size range, new electronic, optical, chemical, magnetic or even
mechanical properties arise due to the effects of the quantized nature of the electronic interactions. Thus, as far as
the development of the novel materials is concerned, the properties of these particles are not predictable by scaling
laws. In this size range every atom counts, i.e. adding or removing just a single atom from the particle will result in
different materials properties. One of the most prominent examples of this kind are the materials formed by
condensing C 60 or related fullerenes into solids bound by van der Waals forces. These purely carbon based solids
exhibit distinctly different properties from diamond or graphite, the other known solid carbon materials.
LCLS offers the unique opportunity to study both the electronic properties as well as the actual atomic structure
of individual mass selected particles with an exactly defined number of atomic constituents. The pulse properties of
5

the LCLS source are required for these studies, since cluster production coupled with mass selection results in a
highly diluted target density. The electronic properties will be accessible by studying the Auger spectrum following
the creation of a core hole, whereas the atomic arrangement will be studied by core level photoemission (XPS) and
(N)EXAFS, once the photon energy at LCLS can be readily tuned. For non-resonant Auger spectroscopy the exact
photon energy does not need to be determined, as long as it is sufficiently high above the core electron absorption
threshold. For XPS the monochromator running in spectrograph mode is required, since the photon energy and
photon lineshape need to be recorded for each LCLS pulse individually together with the XPS spectrum.
Initially in these experiments we will concentrate on the characterization of particles with controlled magnetic
and/or chemical properties.
Control of the magnetic properties of nanoparticles, The control of the properties of magnetic nanoparticles is
crucial for applications ranging from to data storage to cancer treatment. Under basic science aspects, it is intriguing
to note that all atoms, except the rare gases, exhibit magnetism (Hund’s rule). Thus, in the form of well defined
clusters, even elements such as Al form magnetic particles. As extended solids, on the contrary, only the latter part
of the transition metals and the f-electron systems are magnetic. In general, the magnetic moments of clusters are
larger than in the corresponding solids, whereas the temperature, below which magnetic order is stabilized, is lower.
In summary, there are many degrees of freedom to design the magnetic properties of nanoparticles by controlling the
exact size and composition, whereby not only metals, but also oxides are interesting candidates.
Control of the chemical properties of nanoparticles Not only transition metals and their oxides are interesting
candidates as catalysts, but also small Au and Ag clusters and their oxides exhibit quite a high catalytic activity, for
example in various oxidation reactions. Again, the electronic structure of individual mass selected nanoparticles will
be probed by (resonant) Auger spectroscopy and XPS to elucidate not only the particle properties but also the
chemical bonding of adsorbates, while pump-probe spectroscopy will give insight into photo-chemical reaction
cycles.
2.3 Magnetic Imaging
One of important topics of physics is the study of phase transitions, where structural, electronic, or magnetic
properties undergo discontinuous or continuous changes. Powerful symmetry and statistical concepts have been
developed to describe such phase transitions, and corresponding scattering and thermodynamic measurements have
been used for experimental verifications. Such approaches are quite successful and satisfactory to many.
Complementary and equally powerful, the conceptualization of a simple physical picture in the real space has played
an important role in the formulation of physical understanding of many phenomena. This method of direct
representation, however, has not received equal recognition perhaps due to the lack of corresponding experimental
techniques until now. By combining the powerful Fourier Transform Holography (FTH) imaging technique with
LCLS' high brightness, short pulse structure, and fully transverse coherence, the dynamics of magnetic fluctuations
and magnetization relaxation processes can be visualization at extremely fast time scales and at nm resolutions.
This soon-to-be-established new capabilities will not only provide direct experimental proof of the symmetry and
statistical concepts in magnetic phase transitions, but will also have far-reaching impact beyond magnetism in the
studies of critical phenomenon such as other order/disorder transitions or the demixing of binary alloys.
6

Figure 0-2. Illustration of X-ray Fourie Transform Holography experiment.
The proposed experiments will de designed to investigate the critical fluctuations occurring at magnetic phase
transitions. An initial LCLS experiment will demonstrate the existence of magnetic spin blocks by taking snap shot
pictures on a time scale faster than the fluctuations. The next step will be to study the spin block dynamics by
recording a series of such pictures at well-defined delay times. For imaging of the magnetic spin blocks a newly
developed lensless imaging technique will be used. These experiments, however, would not be possible without the
LCLS' ultra-bright, ultra-short, and fully coherent x-ray pulses. Key properties and implications of the proposed
experiment are:
Single x-ray pulse imaging: A single LCLS pulse will be sufficient to obtain an image of the instantaneous
magnetic domain structure which is expected to be static on the time scale set by the ultra-short x-ray pulse length
(230 fs), even close to the phase transition.
50 nm spatial resolution or better: It has been demonstrated that a 50 nm spatial resolution can be achieved
with the newly developed lensless x-ray imaging technique which will take full advantage of the unprecedented full
coherence of LCLS x-ray pulses. Subsequent phase retrieval may improve this to near-wavelength limited spatial
resolution.
Dynamics of critical fluctuations: To image the fluctuation dynamics we will split the beam and produce
consecutive x-ray pulses with a well defined time separation at the sample. From each pulse we will obtain an
image of the magnetic domain structure, using a scheme discussed below, and thus resolve the dynamics occurring
on a femtosecond to nanosecond time scale.
Ultrafast, non-deterministic dynamics: It is important to realize that the proposed experiments significantly
differ from today’s ultrafast pump-probe experiments. Such experiments rely on reversibility of the sample to a well
defined state before each pump-probe cycle. In contrast, the critical fluctuations that are the subject of our study are
non-deterministic and their study requires complete images to be recorded in a single shot.
Ultrafast relaxation dynamics: It is clear that once we have demonstrated the feasibility of ultrafast single
shot imaging a whole class of new experiments will become possible. Besides single shot imaging of spontaneously
occurring fluctuations we also envision to study ultrafast relaxation dynamics in pump-probe experiments.
7

2.4 X-Ray Scattering Spectroscopy on Strongly Correlated Materials
Among the current research issues of condensed matter physics, the electronic structure of the strongly correlated
materials is one of the most active topics. In this class of materials, the Coulomb interaction between electrons can
not be ignored, which manifests itself as “strong correlations” acting as a “tuning parameter” to switch the ground
state from one to the other. These collective ground states for different phases are known as “emergent” phenomena
of many body systems, which can not be deduced from any perturbation theory and often involve novel forms of
order.
X-Ray scattering experiments on strongly correlated materials have been performed to probe the charge ordering
and elementary charge/magnetic excitations containing important information of the ground state properties. Recent
theoretical developments show that resonance x-ray scattering can provide rich information on many-body
wavefunctions. Soft x-ray, being sensitive to valence electrons and in the spectral range of important L edges of
transition metals that often are key elements of correlated materials, provide special opportunity. However, so far
most of these experiments were performed in the equilibrium state, which can not provide any information along the
time axis regarding how the electrons form this particular ground state. The exciting opportunity provided by the
SXR of LCLS is exactly this missing piece of information along the time axis. Using the high pulse intensity and
ultra-short pulse length, it is possible to perform optical-pump-and-X-ray-probe experiments to study how the
electronic states of strongly-correlated materials relax from an excited state to the ground state. The relaxation
process is closely related to the correlation effect among the electrons and the electronic interactions to other degrees
of freedom; therefore “snap shots” obtained from the pump-probe experiments provide important clues to construct
a microscopic physics picture of the strongly-correlated systems. In addition, x-ray probe experiment also has some
unique advantages, such as element specific information, bulk sensitive signal, and the dynamic structure factors,
which are not accessible by the most common ultrafast optical pump-probe experiments in the visible light regime.
Figure 3 (a) A sketch of the momentum-resolved inelastic scattering experiment chamber. (b) The momentum
transfer covered by different locations of the spectrograph at the Mn L edge. (c) The Fermi surface of La 1-
xSr 1+x MnO 4 and the region which can be covered by this spectrograph with the designed rotary sample stage.
(d) The dispersion of orbital ordering of the La 1-x Sr 1+x MnO 4 .
Three kinds of experiments were proposed for using LCLS to study the physics of the strongly correlated materials:
Absorption experiment: Absorption spectrum reveals the partial density of states of the unoccupied state of
the materials. As a first step, it is important to understand how density of state change after the system been
“pumped” by the optical Laser pulse. It is also an ideal initial experiment to do for the initial operation stage of the
LCLS, since it is not an extremely photon hungry experiment, which can be done using with less powerful pulse and
lower repetition rate. In addition, this absorption experiment would become a routine diagnosis experiment for the
chamber alignment, sample damage assessment and a survey experiments for the resonant scattering experiments.
8

Resonant Elastic Scattering experiment: Charge, spin and orbital order is one of the interesting phenomena in
many strongly correlated electron systems. These orderings, mostly incommensurate to the lattice constant, produce
extra Bragg peaks in the elastic X-ray diffraction pattern. As these orders are often complex and thus relatively
large in real space, making it possible for soft x-ray to have sufficient q to probe the extra Bragg peak. Using SXR
pump-probe capability of LCLS, one could destroy the charge ordering by the optical pump Laser and probe by the
LCLS X-ray pulse at different delay times after the pump as the system relaxes back to the charging ordering state.
This experiment shall reveal important microscopic information of the charge ordering formation.
Resonant Inelastic X-ray Scattering (RIXS) Experiment: To probe the excitations of the ground state, it is
necessary to record the energy loss of an inelastic scattering process. In addition, the information at different
momentum transfers of the scattering process is also extremely important for the scattering experiments on solids as
it is related to the momenta of these excitations. Time resolved pump-probe RIXS experiments can be used to
measure time evolution of collective modes and thus, information about the dynamics involves in the many-body
excitations that gave the collective modes. In addition, we will also use RIXS to obtain the wave function properties
through various projections to its intermediate states. The pump-probe experiment using the SXR shall provide a
description of the wave function evolution from the pumped exciting state to the ground state.
2.5 High-Resolution Ultrafast Coherent Imaging
X-ray microscopy at synchrotron sources is steadily progressing, with the nanofabrication of better zone-plate
lenses and the development of lensless coherent imaging techniques. However, even with cryogenic sample
cooling, radiation damage limits the achievable resolution to about 10 nm. Higher resolution is required to
understand the structure and organization of living (unstained and unsectioned) cells, and would greatly complement
real-time optical fluorescence microscopy to study cell processes, such as cell division, and full function of
components such as the cytoskeleton. Coherent diffractive imaging with intense and ultrashort X-ray pulses could
achieve the required resolution on living cells by recording the scattering information before any structural changes
due to interaction with that pulse. This method of flash imaging has been verified at FLASH. In principle the
resolution should only be limited by the wavelength of the radiation, given the appropriate pulse parameters and size
of the focused pulse. The details of the matter–FEL interaction must be studied to gain an understanding of
achievable resolution limits. Methods must also be developed for 3D imaging of reproducible samples, using
streams of particles in the gas phase or in droplets. These delivery systems, first developed at FLASH, are required
to quickly replenish the sample and allow time-resolved stroboscopic imaging of laser alignment of particles. Also
the coherent diffractive method (with fixed or injected samples) enables the highest spatial resolution of ultrafast
processes in non-periodic systems, such as the study of laser-induced phase transitions in materials. Experiments at
FLASH demonstrated this technique on the study of phase separation in laser-ablation, but the shorter wavelengths
of LCLS are required to achieve the necessary spatial resolution.
Imaging of biological cells beyond radiation damage limits: With the X-ray intensity provided by 5-micron
focusing it will be possible to achieve a single-shot resolution of better than 5 nm. Imaging will be carried out on
cells on thin membranes that can be placed into the beam as well as on cells injected into the LCLS beam.
FEL-matter interactions: Initial experiments will study the effects of damage by collecting coherent patterns
of homogeneous samples (such as polystyrene spheres or nanocrystals) as a function of pulse fluence. In this case,
the scattering pattern immediately gives the pulse-integrated size distribution of the particles, which can be
compared with theory. Nanocrystals of biomolecular complexes (such as PS1) will give the high-resolution
information about the damage of organic material as a function of resolution and pulse fluence.
Time-Delay Holography: This method, tested at FLASH, will be implemented to measure movies of the
interaction and evolution of reproducible samples (such as virus particles) with FEL pulses. The apparatus will
include a reflecting crystal, such as InSb, to direct the pulse back onto the sample. The detector will be placed in a
9

ackscattering geometry. Samples will include layered spherical structures to test methodologies to prolong the
onset of damage.
Laser-matter interactions: A synchronized optical pump pulse will be used for single-shot time-resolved
imaging in materials and stroboscopic imaging of laser-particle interactions (such as laser alignment).
10

3. Science Driven Requirements:
In this section the basic science driven requirements of the beam line, optics and experimental stations for the SXR
system are described. The proposal is for a monochromatic soft x-ray beam line with two positions for experimental
stations.
3.1 Experimental stations:
The first experimental station shall be up stream of the monochromator in the unfocused beam. The second after the
monochromator for focused beam.
3.1.1 Experimental station position 1:
It is important to allow for an experimental station at a position before the monochromator where x-ray interactions
could be performed in a single shot mode. This is to be a simple chamber to allow transmission experiments on large
samples, > 1mm in diameter. The requirement for absorption spectroscopy is a detector at the monochromator exit
slit. A pump laser is required for pump probe experiments at this position. Ultra-high vacuum,

3.2.3 Pulse stretching:

3.5.1 Pump Laser:
The pump laser should be capable of producing 2mJ/pulse at 120Hz. The pump laser system is defined in Physics
Requirements for LCLS/NEH Laser System, Requirements Document # 1.6-010 rev 0 section 2, 3, 4, 6, 7 and 8. To
summarize it should be capable of delivering 800nm with an energy of up to 3mJ in a pulse ≤50 fs at the sample.
The pump laser shall be synchronized with the LCLS FEL at a repetition rate of 120Hz. The pulse laser should be
synchronized to the LCLS RF with a jitter of ≤100 fs. The delay system should be capable of adjusting the laser
timing by ±1ns with 10fs accuracy. There are some experiments that require 25 mJ/pulse at a 10Hz rate. The
conversion to shorter and longer wave lengths is not required for initial experiments on the SXR (section 5 of the
above mention document), but provision should be made to accommodate this at a later date.
3.5.2 Alignment Lasers:
A HeNe alignment laser beam shall be set up coaxial with the FEL beam to facilitate the positioning of samples in
the FEL beam. This laser shall be class IIIa rating or less.
13

4. Optical Design
4. 1 Optical layout
At synchrotron radiation sources, two of the most successful types of grating monochromators are the
Varied-Line-Spacing grating monochromator (VLS) and the Plane Grating Monochromator (PGM). The VLS
monochromator was developed by Underwood, Koike and coworkers at the Advanced Light Source (ALS). [2] A
spherical mirror produces a converging beam in the vertical plane and a varied-line-spacing grating diffracts the x-
rays onto an exit slit. The variable period of the grating provides additional parameters to keep the focal distance
constant as a function of photon energy and to compensate for aberrations of the mirror. The focal plane is erect,
which is convenient for implementing the spectrometer mode. The photon energy is scanned by a single rotation of
the grating.
The PGM monochromator was developed by Peterson, Follath and Senf at BESSY. [3] In the PGM
monochromator a plane mirror reflects the x-rays onto the grating and an elliptical or spherical mirror focuses the
beam onto the exit slit. The plane mirror provides a variable included angle on the grating, which allows the grating
efficiency to be optimized over a wide photon energy range. The photon energy is scanned by coordinated rotations
of both the plane mirror and grating.
We have chosen the VLS monochromator type because of its simplicity. Only two optical elements are
required, and only the grating rotates. The limited photon energy range of the SXR Instrument, 500 to 2000 eV,
would not take advantage of the variable included angle of the PGM monochromator.
Figure 4-1. Optical layout of the SXR Instrument.
The optical layout of the SXR Instrument is shown in figure 1. Table 1 is a list of the optical elements and
their parameters. The first component is experimental chamber 1, which is a location for samples in the unfocused
non-monochromatic beam. Next, there is the M1 spherical mirror and G1 and G2 plane gratings of the VLS
monochromator. In monochromator mode, an exit slit selects a narrow bandwidth. Alternatively in spectrometer
mode, a detector will measure the dispersed x-ray absorption spectrum. The M2 plane elliptical mirror provides the
horizontal focus in experimental chamber 2. The plane elliptical M3 mirror produces a vertical image of the exit slit
at experimental chamber 2. The experimental chamber 2 provides the experimental environment for monochromatic
and focused pink beam experiments. Endstation 1, the M1 mirror and gratings are in the first hutch of the LCLS
Near Experimental Hall. The rest of the beamline from the exit slit and detector through the experimental chamber
2 are in the second hutch.
4.2 Optical elements
The design of the optical elements adopts characteristics of the LCLS soft x-ray offset mirrors in the Front
End Enclosure. The incidence angle of the mirrors is 89.14o equivalent to a grazing angle of 15 mrad. The mirror
substrates are single crystal silicon, and the preferred optical coating is B4C. Figure 2 shows the reflectivity of B4C
in comparison with other coating materials. The reflectivity of B4C is excellent about 90 % over the whole energy
14

Type
Exp.
Chamber 1
M1 Spherical
mirror
G1, G2 Plane VLS
grating
Detector/
Slit
M2 Bent Elliptical
mirror
M3 Bent Elliptical
mirror
Exp.
chamber 2
Coating and
blank
material
B 4 C-coated
silicon
B 4 C -coated
silicon
B 4 C-coated
silicon
B 4 C-coated
silicon
Dimensions
l x w x t
(mm)
Clear
Aperture
(mm)
Radius
(m)
Figure
error
(μrad)
Roughness
(nm)
Incidence
angle(°)
Grating period
order
Distance
from
source (m)
124
250 x 30 175 x 10 943 0.3 0.4 89.14 - 125.1
220 x 50
x 23
170 x 24 ∞ 0.5 0.4 88.7-88.9 1/100, 1/200
-1
250 x 30 190 x 10 262.8 0.3 0.4 89.14 - 137.4
250 x 30 125 x 10 156 0.4 0.4 89.14 - 137.9
125.4
132.6
139.4
Table 4-1. The optical elements of the SXR instrument.
15

1.0
0.8
Reflectivity
0.6
0.4
0.2
B 4
C
C
Si
B
0.0
5 6 7 8 9
1000
2
Photon energy (eV)
Figure 4-2. The reflectivity at 15 mrad grazing angle of B 4 C, C, Si and B.
range of the SXR instrument. Boron and Carbon would also give acceptable performance. The incidence
angle and coating materials maintain 2000 eV as the high photon energy limit. Silicon would limit the cut off to
1800 eV. The clear apertures are set to accept a 5 σ footprint of the x-ray beam including a tolerance of 0.5 μrad
rms for the LCLS beam pointing stability. [4] The M1 mirror is polished as a sphere. On the other hand the M2 and
M3 mirrors are polished as flats and then bent into their plane elliptical shapes. The M2 and M3 mirrors need an
elliptical surface in order to eliminate the coma aberration. The coarse grating line densities, 100 and 200 l/mm, are
a consequence of the included angle 2 θ being close to 180 o . It is planned to have both rulings placed on a single
grating substrate. The gratings operate in the negative first diffracted order. The choice of negative order has two
benefits. The larger grazing incidence angle reduces the required length of the grating. The grating in negative
order has magnification, which eases the needed detector spatial resolution. For the distances of the optical elements,
the source position is chosen as 10 m upstream from the end of the undulator. [5]
Figure 4-3a. The LCLS FEL source at 1000 eV.
Figure 4-3b. The x-ray beam at experimental
chamber 1.
16

Figure 4-3c. Photon energies 999.8, 1000 and Figure 4-3d. The monochromatic focus in
1000.2 eV at the detector or exit slit. experimental chamber 2.
4.3 Raytracing, energy resolution and efficiency
Raytracing has been performed to confirm the optical design using the XOP software. Spot
diagrams are displayed in figure 4. The LCLS FEL source at 1000 eV is predicted to be round with a
diameter of 82 μm (fwhm) and a divergence of 8 μrad (fwhm). Compared with third generation
synchrotron sources, the horizontal beam size and the divergence in both dimensions are significantly
smaller. In experimental chamber 1 the unfocused x-ray beam is again round with a diameter of 1 mm
(fwhm). The M1 mirror and VLS gratings G1 or G2 produce a vertical focus, 1.1 mm horizontal by 18 μm
vertical (fwhm), at the exit slit or detector. The spot diagram (3c) shows three different energies 999.8,
1000 and 1000.2 eV at the exit slit. That these three photon energies are well resolved confirms that the
resolution goal of 0.2 eV at 1000 eV is achieved. The M2 and M3 mirrors refocus the x-rays horizontally
and vertically into experimental chamber 2. The predicted monochromatic focus in experimental chamber
2 is between 1 and 2 μm horizontal by 3 μm vertical (fwhm). In the case of the non-monochromatic beam,
the vertical focusing is changed because now the grating has a magnification of unity. For the nonmonochromatic
beam the calculated focus in experimental chamber 2 is nearly round between 1 and 2 μm
in diameter (fwhm). It should be noted that this ray tracing does not include optical fabrication errors.
The tolerances for FEL x-ray optics are quite demanding and beyond what is required for
synchrotron radiation optics. This difficulty can be simply understood by the fact that the FEL and
synchrotron source dimensions are similar but for FELs the first optic is roughly ten times further away;
hence the allowable slope error is reduced by about an order of magnitude.
s
2Δ r ≤ , (1)
T
2
where Δ T is the tangential slope error, r the source distance and s the source dimension. Preserving the
source brightness of the LCLS is challenging.
Optical tolerances are included in table 1 for the slope error and roughness. The figure
specifications for the M1 mirror and G1 and G2 gratings result from the energy resolution goal of 0.2 eV at
1000 eV. The figure tolerance for the M1 mirror is the most difficult because the image created by the M1
mirror is magnified by the grating. The figure specification of the M2 and M3 mirrors is derived from the
required focus in experimental chamber 2, a 10 μm diameter. The B 4 C coatings place an upper limit on the
roughness for spatial periods from 20 nm to 2 μm, the Atomic Force Microscope measurement range. With
higher substrate roughness the B4C coating growth process changes. A specification from Fourier optics
considerations will be added to the mirror and grating specifications.
17

The grating efficiency is the most important factor in determining the overall beamline efficiency.
Efficiency calculations were performed with the GSolver code. [6] The optical constants for the B 4 C
coating were taken from the CXRO website. [7] For the laminar groove profiles, the groove depths and
widths were varied to maximize the efficiency at single photon energies: 800 eV for the 100 line/mm
grating and 1200 eV for the 200 line/mm grating. The optimal groove depths were found to be 19 nm for
the 100 line/mm and 13 nm for the 200 line/mm grating. High peak efficiencies between 0.1 and 0.4 were
calculated. These efficiencies should be related to the low groove densities and the high reflectivity of B 4 C.
The average power in the LCLS FEL radiation is low, 0.2 W, because of the low repetition rate,
120 Hz. On the other hand, the peak power is quite high, 5 GW, as a result of the ultrashort pulses. Optical
damage from the LCLS x-rays was modeled by London et al. [8] Their guideline is to stay below the melt
fluence. The risk of optical damage is reduced by using small grazing angles and low Z coatings. Since the
SXR mirrors employ the same incidence angle and coating as the LCLS soft x-ray offset mirrors, these
mirrors should be safe from the damage calculations of these mirrors. An estimate of the absorbed energy
for the M1 mirror is 0.04 eV/atom, well below melting dose of 0.62 eV/atom. The case of the gratings
could be worse because of the larger grazing incidence angle and that a portion of the x-rays strike the
leading edges of the lands at near-normal incidence. However, estimates of both these cases gave an
acceptable dose of 0.05 eV/atom. These damage calculations were performed at 2000 eV, which is the
most difficult case because of the reduced beam divergence. The optics of the SXR instrument should not
damage from the FEL radiation.
The exit slit may be also damaged from the LCLS FEL beam, which is vertically focused at this
location. Three cases must be considered: monochromatic beam, zero order and pink beam. For the
monochromatic situation the dose is acceptable, 0.03 eV/atom, because the intensity is reduced by the
dispersion of the different FEL wavelengths in negative first order. This damage consideration does
require that the exit slit blades be made of B 4 C or perhaps another low Z material. The zero order can be
blocked at a small distance downstream of the gratings, where the x-ray beam is not focused. The final and
most difficult case is the undispersed pink beam. Here the dose at the exit slit, 5 eV/atom, is not acceptable.
One solution is to add an aperture upstream, where the FEL beam is large enough and the flux density
below the damage limit. This aperture would need to be aligned to the exit slit and the exit slit would need
to be openable to a larger width than the aperture. An aperture that opens to 2 mm would transit the
dispersed spectrum to the detector for spectroscopy experiments.
Grating optics increase the x-ray pulse duration. This pulse stretching, Δt, results from the extra
optical path, m λ, between groove n and groove n+1, where m is the order of diffraction and λ the x-ray
wavelength. It can be calculated from
t =
Nmλ
c
Δ , (2)
here N is the number of illuminated grooves and c the speed of light. For 800 eV the estimated pulse
stretching is 40 fs, which is well below the predicted LCLS pulse duration of 320 fs. For measurements
with the sample in experimental chamber 1 or in experimental chamber 2 using pink beam, there is no
optical pulse stretching. There are operating modes of the LCLS which will provide shorter pulses. [9]
There should be an angular aperture in the SXR instrument in order that the pulse stretching can be reduced
with a corresponding loss in intensity and in energy resolution.
There are existing mechanical designs for monochromators, bendable mirror assemblies and exit
slits at the Advanced Light Source and other synchrotron facilities. It is planned to reuse existing
mechanical designs with minimal modifications for the LCLS SXR instrument. Switching between
monochromatic and pink beam may be accomplished in a convenient manner provided by the current,
18

“standard” ALS monochromator chamber design. The chamber is translated horizontally perpendicular to
the beam propagation direction. If the grating surface has ruled and unruled areas, the grating can either
diffract for monochromatic operation or reflect for pink beam operation.
References
1. P. Emma, private communication.
2. Underwood, J.H., and. Koch, J.A, Applied Optics 36, 4913-4921 (1997).
3. H. Petersen, Opt. Commun. 40, 402-406 (1982); R. Follath, Nucl. Instrum. and Meth. A, 467, 418-425
(2001).
4. P. Stefan, private communication.
5. Y. Feng and H. D. Nuhn, private communications. The source distance depends upon the point along the
undulator length where the FEL reaches saturation. 10 m from the downstream end of the undulator is the
nominal saturation point for the LCLS start up parameters at 800 eV.
6. http://www.gsolver.com.
7. http://henke.lbl.gov/optical_constants.
8. R.A. London, R.M. Bionta, R.O. Tatchyn and S. Roesler, SPIE Proceedings 4500, 51-62 (2001).
9. P. Emma et al., Phys. Rev. Lett. 92, 074801 (2004).
19

5. Instrument Layout:
In this section we will describe the constraints on the physical layout of the SXR systems and how we
propose to meet the scientific requirements within those constraints. The proposal conforms to the new
AMO experiment location in the first hutch of the Near Experimental Hall (NEH) on the 83 mrad line. The
SXR systems are on the 28 mrad line, between the AMO line and the hard x-ray line, with a premonochromator
sample position, M1 mirror and grating system in first hutch. The SXR exit slit,
refocusing mirror and end station are in the second hutch.
In the Front End Enclosure (FEE) the M3-S1 or M3-S2 mirrors of the Soft x-ray Optics Mirror System
(SOMS) direct the FEL beam down either the 83 mrad or the 28 mrad lines to the AMO and SXR
experiments respectively. On each branch there are fixed collimators and insertable photon stoppers in the
FEE that block the beam, so entry can be permitted into the first hutch. Just down stream of the first hutch
wall there is an insertable beam position imaging system and an isolation valve. This terminates the SOMS
vacuum system.
The basic optical layout is presented in section 3. There are several constraints on fitting this optical
design into the NEH. The first is that it does not materially interfere with the program on the AMO
experiment. Both requirements are physical, over space, and operationally for access during operations on
each system. A second is the flux in the FEL beam. Though power is on average low, the flux density in
the FEL beam will damage most materials. This is particularly critical where the beam is focused. The third
requirement is the inclusion of a moderately high resolution monochromator, which essentially requires it
span the first and second hutches.
The AMO group is locating their experiment in the first hutch of the NEH. The SXR beam will pass
through the first hutch, horizontally between the AMO experiment and the hard x-ray beam line. The first
experimental station and the monochromator would be located in the first hutch just up stream of the AMO
focusing optics. These would have to be properly shielded for personnel to be in the first hutch while beam
is passing through the SXR monochromator into the second hutch. These shielding requirements have not
yet been defined. The controls on the shielding will follow SLAC personnel protection requirements. This
shielding and controls will be required for the SXR experiment to run effectively. The hard x-ray beam line
that also passes through the first hutch will be shielded.
The flux density in the FEL beam will damage most materials and is particularly critical where the beam is
focused. Materials of low Z elements have the best properties for surviving in the beam [3]. Collimators
and beam stops are to be fabricated from B 4 C which should not damage in the unfocused FEL beam at this
distance from the source. At approximately 0.12 eV/atom/pulse fatigue sets in and at 0.6 eV/atom/pulse
single shot damage occurs in B 4 C. The fatigue value sets a limit for optics and beam stops that are
continuously or often in the beam. Collimator and apertures that will only see incidental beam strikes are
limited at the higher single shot damage threshold. The monochromator proposed in section 3 focuses, in
one dimension, both the monochromatic and the zero order, white, beam at the exit slit. All materials are
likely to damage in the focused zero order beam at the exit slit. Thus the exit slit must be opened when
zero order light is put through it to prevent damage to the slit blades. This does not substantially limit the
ultimate focus at the second experimental station as the zero order focus is quite small, 1σ ~10μm. An
aperture up stream of this zero order flux density ‘stay clear’ would contain the beam and prevent the
focused zero order beam from striking the open exit slits.
5.1 Instrument Configuration:
The SXR proposal is predicated on the inclusion of a monochromator, which essentially requires the
optical system to span the first and second hutches of the NEH. The basic layout is shown in figure 5.1. The
transmission sample position and the monochromator focusing/grating system will be located in the first
hutch along with, but up stream of, the AMO experiments. The SXR is laid out to make space for an
extension of AMO experimental systems into the second hutch.
20

Figure 5.1 Proposed layout of the AMO and SXR beam line and experimental systems in the NEH. SXR in
orange, AMO in purple and XTOD, to down stream hard x-ray experiments, in blue. (Dimensions in m.)
The hutches are roughly 10m in length , 10m in width and 4.5m high. Access is through interlocked doors.
The hutches will also contain the optical tables for the pump and alignment lasers, electronics racks
electrical panels and other utilities. See figures 5.2 A & B. The first hutch is primarily dedicated to the
AMO instrument on the 90mrad branch line (actually 83 mrad from the undulator center line). On the 30
mrad (23mrad) branch line is the SXR instrument. In the first hutch the single pulse shutter, the
transmission sample chamber, the monochromator tank, collimators and drift tubes for the SXR will be
located. On the 0 mrad line the XTOD, hard x-ray beam line, passes through both hutches 2m from the
back wall. The nominal FEL bean height is 1.4m above the floor.
The SXR optical layout has the M1 mirror and grating system just up stream of the AMO experiment
which puts the exit slit, 7.5m down beam, just inside the second hutch. The K-B pair of refocusing mirrors
are in the middle of the hutch with the final focus 1.5m down stream of the center of the M3 mirror. See
figure 5.2 A and B. The focus is 1.3m from the hard x-ray beam pipe and 3.3m to the back wall of the
second hutch. The XTOD beam pipe is shown with a n0.1m beam pipe, but additional space will be
required for valves, stands and shielding. The shielding requirements have not been defined at this time.
The location of stands and valves is some what negotiable with the XTOD if done early on. The XPP
experiment going into the third hutch requires space for a slit system in the second hutch. The proposed
location is just up stream of the K-B mirrors. There will be electronics racks along the back wall of the
hutch and access to these racks much be provided. A ~1m (36”) access way around the end of the
experimental system is shown.
From the focus there is ~2.4m down beam and ~1.2m on XTOD side available for experiment systems. The
flange to the beam line is laid out at 0.5m before the focus. There is an optical bench for the pump laser
near the center of the hutch. The pump laser beam is likely to be a bit lower than the FEL beam height and
then introduced by directing it up into the vacuum system The largest system proposed so fare,
Down beam of the AMO experiment there is space marked out for a future extension of the AMO beam
line. This station could be configured in several ways. This area should be kept clear of any permanent
installations.
21

Figure 5.2 A: layout of AMO and SXR instruments in the first hutch.
Figure 5.2 B: Layout of SXR experiment in second hutch with potential third AMO experimental station.
22

5.2 Single Pulse Shutter System:
The first item in the SXR line is to be a Single Pulse Shutter (SPS) system. The system is to be a clone of
the shutter system in development for the AMO experiment. The fast shutter system can close or open in
less than 8 ms The solenoid driven shutter will be triggered through the EPICs control system (see controls,
sec 5.3). The position of the shutter will be adjusted with a motor driven manipulator and a camera will be
mounted to observe the beam on the shutter’s up stream face for alignment. The shutter is essentially
binary, it will be either in the beam of out of it.
5.3 Transmission Sample Station, Pre-Monochromator Position 1:
There will be an experimental station before the monochromator, experimental position 1, where samples
can be introduced into the unfocused , ~1mm in diameter, beam. Several detection methods are potentially
possible. The proposal is to concentrate on transmission through thin samples, using the monochromator in
spectrograph mode, with a detector at the exit slit to measure absorption spectra.
The absorption experimental set up will be simple. A x-y manipulator will be provided to position the
samples in the beam. Configuration of the sample holder and sample introduction system has yet to be
defined. The sample holder will include a Ce doped YAG crystal for viewing and alignment of the FEL
beam. There will be a camera to monitor sample position and to align the FEL, the pump laser into
overlapping phase space at the sample position. The camera will be capable of 120Hz imaging and
synchronized with the FEL pulse and time stamped. The spectrometer detector, at the exit slit, will fully
instrumented and capable of operating at 120Hz with the data time stamped to each pulse.
The sample chamber will be isolated from the SOMS and the monochromator optics by a gate valve. The
chamber will have an ion pump for UHV operation and a cold cathode gauge to monitor the pressure. The
pressure in this system will have to be

5.4 Monochromator:
The AMO systems occupy the later two thirds of the first hutch. This is to maximize the space between the
28 and 83 mrad beams. It opens up a limited area in the up stream end of the first hutch for an
experimental position and the M1 mirror and grating system on the 28 mrad/ SXR line. See figure 5.1.
The first focusing element in the SXR beam is the M1 mirror. This mirror will have a fixed spherical
radius. It will focus the beam at the exit slit of the monochromator. The angle incidence on the mirror can
be adjusted to get the focus at the exit slit. The mirror pitch will be motorized and must have the range to
both compensate for the as delivered radius of the mirror and correct for changes the virtual source distance
as the FEL parameters are varied. The pitch motion will have limits and hard stops set so the beam can only
strike the surfaces designed to take the FEL flux. The mirror will vertically deflect the beam upwards by 30
mrad.
The grating immediately follows the M1 mirror. It will be a Variable Line Space (VLS) type grating on a
flat substrate. With this VLS grating the focal length of the monochromatic radiation is constant and at the
same distance as the zero order, specularly reflected, radiation. The monochromator therefore has a fixed
exit slit and can easily be operated as a spectrograph by placing a position sensitive detector at, or near, the
exit slit. Both the M1 mirror and the grating will be fabricated from single crystal silicon and have B 4 C
optical coatings [2]. The grating will vertically deflect the beam downwards with the output parallel to the
input beam.
Monochromator mechanisms and optical elements are commercially available. The outline of such a
commercial system, which meets the requirements of section 3, shown in figure 5.1.
The grating pitch and exit slit aperture will be interlocked to prevent damage to the exit slit blades. The
pitch motion of both the M1 and gratings will have limits and hard stops set so the beam can only strike the
surfaces designed to take the FEL flux.
5.5 Zero Order Stop and Grating Aperture:
Down stream of the grating there shall be a photon stop that will absorb the specularly reflected radiation
from the grating when it is operating with diffracted beam going down through the exit slit. The
monochromator is designed to only operate with out side orders, therefore the zero order stop will be on
lower side of the transmitted beam. The zero order stop will extend passed the limits of beam travel when
both the M1 and grating are at there maximum pitch. As there will be beam on the zero order stop when
running in monochromatic or spectrographic mode the zero order stop will be either configured to be either,
safe for indefinite exposure to the beam or back by a Burn Through Monitor.
Down stream of the zero order stop and up stream of the exit slit collimator will be an adjustable vertical
aperture that can be used to limit the number of lines on the grating that can illuminate the exit slit. This
will be used to limit the temporal stretching of the pulse, at the expence of flux and resolution. (See section
3.)
5.6 Exit Slit Collimator:
There will be an collimator between the zero order stop and the exit slit. This collimator has to be upstream
of the zone in which the focused zero order beam can damage B 4 C. This collimator will have an aperture
that is smaller than the exit slit when it is fully opened. This will prevent zero order beam from striking the
exit slit, which is interlocked to the grating pitch as described below. The outer diameter of the collimator
will be sufficiently large to contain the beam.
24

The SXR beam pipe will pass the AMO systems between the grating system and the exit slit. The only
components in the SXR line adjacent to the AMO system would be apertures and diagnostics up stream of
the zero order stay clear and these could be located so as not to interfere with AMO components. The
collimator and beam pipe to the exit slit shall be shielded so personnel can work in the first hutch when
beam is going through SXR monochromator.
5.7 Exit Slits:
The Exit Slit will be located just inside the second hutch as shown in figure 5.1. The exit slit shall open to a
gap larger than the up stream collimator and close to

5.10 Intensity Monitors:
The spectrum as well as the intensity of LCLS will fluctuate due to the SASE process. LCLS provides
measurements of the number of photons on a shot by shot basis. As the spectrum of LCLS fluctuates
beyond the spatial resolution of the monochromator, it is essential to measure the beam intensity after the
slit assembly on a shot by shot basis to normalize measured spectra. The intensity monitor need only be
relative and linear over a short energy range, on the order of 10 eV. A signal to noise ratio of 0.1% is
needed.
A gas cell can exceed these requirements. Such a cell should be windowless to preserve the coherence of
the FEL beam. such a cell is complex to design and expensive construction and is not in the scope of the
first experiments. Space between the slit assembly and the first focusing mirror tank should be kept clear as
far as possible for future use by a future gas cell monitor.
The use of an in-vacuum photo diode to measure the fluorescent and scattered radiation from an optical
element will be implemented inside the second focusing mirror tank. Similar intensity monitors have been
implemented successfully at beam lines at SSRL. Scaling from the SSRL data and assuming 1*10^12
photons per pulse, there would be 0.5x10^6 photon per pulse in the diode, at 800 eV this would be 1*10^8
electrons per pulse which should approximately 0.15% signal to noise. For better statistics either the diode
could be moved closer to the mirror, i.e. make detector more position sensitive, or use several diodes, to
make it less position sensitive. As these detectors are only collecting scattered radiation from optics, they
are non-destructive to the beam. Such detectors can configured specifically to be beam position monitors.
The photocurrent from a thin Aluminum foil has been implemented successfully at FLASH to be used as an
intensity monitor for the fifth harmonic. The radiation after the monochromator is attenuated by two orders
of magnitude and is not expected to destroy the foil. The aluminum foil assembly will be mounted on a
motorized linear feed-through which allows one to insert and remove the assembly into / from the beam. A
ring electrode at +1kV around the beam separated from the foil by 2mm is used to collect the photo
electrons, avoiding non-linearity caused by space charge effects. The Al foil itself is connected to a charge
sensitive amplifier through a coaxial cable and the detected signal is digitized at 120 Hz, synchronously to
the repetition rate of LCLS. The Al foil will be 200nm thick (commercially available). Beryllium foils are
also an option.
5.11 Refocusing Mirrors:
Refocusing optics are required that image the exit slit vertically, and the source horizontally at the second
experimental position. The configuration from section 3 is to put these mirrors close to the final focus
while being compatible with differential pumping and experimental chamber dimensions.
The proposed imaging system will be a K-B pair of elliptical cylinders. The elliptical cylinders will be
generated by bending optical flats that have a profiled width and independent moments applied at each end.
The mirror system shall be able to adjust from essentially unfocused to fully focused at the sample position.
The mechanical design is to be very similar to that used for the AMO focusing mirrors with the exception
that they do not need to be removed from the beam. The M2 and M3 mirrors are to be operated at
essentially fixed angles of incidence. A fixed aperture will be located down beam of the M3 Mirror to
contain the beam. Thus the components between the M3 and the focus can be fixed in space.
The K-B mirror system will be a clone of the AMO system with the exception that the profiles of the
mirrors will be optimized for the focal lengths on the SXR beam line. The mirrors will be at a nominal
grazing angle incidence of 13.85 mrad and fabricated from single crystal silicon with a B 4 C optical coating.
26

5.12 Experimental End Station, Position 2:
Experimental position 2, the end station, is locate in the second hutch after the K-B refocusing mirrors. The
experimental systems for this second station are not part of this proposal. The SXR proposal provides the
basic requirements of:
• A focus of

5.13 Optical Pump Laser Transport, collinear in-coupling
Optical femto second Laser pulses to excite or probe ultra fast dynamical processes have be guided from
the existing Near Experimental Hall optical laser facility (W. White) to the two sample interaction points,
transmission sample chamber and end station. The pump laser will be transported to optical table adjacent
to in-couple mirrors just up stream of each sample location. Focusing and pointing optics and feedback
will be located out-of-vacuum. This allows for flexibility in bringing the laser beam in coaxially or near
coaxially with the Hamburg in-coupling mirror system, described below, or though an off axis viewport.
The nominal set up will have the laser bean brought in though the Hamburg in-coupling mirror with a
~1mm spot size at the transmission sample position and ≥100μm spot at the end station.
The design is based on the Hamburg design used at the FLASH PG2 beam line. It has 2 inch diameter
dielectric plane mirrors with a central bore to transmit the X-ray beam. For different wavelengths specific
mirrors are mounted on a ball bearing stabilized manual translation stage. At 2 inch diameter and a central
bore diameter of 2 mm minimum distortion of the optical wave front is ensured. In the initial stage 800 nm
and 400 nm mirrors are used. The mirrors are in vacuum mounted in spring mounted holders, which allow
bake out of the UHV system without putting mechanical stress onto the mirror surfaces. On the back of the
mirrors matching fluorescence screens are mounted which allow rapid positioning of the X-ray beam
through the central bore.
In this implementation the mirror chamber is fixed with the pointing feed back done out of vacuum. In the
case of the transmission sample position, with the large spot size the pointing stability is not critical so the
in-coupling mirror and the pointing feed back loops are supported separately. At the end station with the
potentially small spot size the in-coupling mirror and the feedback loops and focusing optics will be
mounted on the same optical table.
Figure 5.5 The operational design at the PG2 beam line at the Free Electron Laser at Hamburg (FLASH)
facility.
28

Appendix C: Descriptions experimental systems groups propose to bring. Scheduling of chambers will
depend on the interest of the community and rankings of the review panel.
29

Dear SXR representative,
This letter describes the intention to install, operate and support a fully equipped endstation at the Soft X-
ray for Material Science (SXR) beamline at LCLS. The endstation will comply with the requirements
outlined in chapter 5 of the SXR Technical Design Report and be avaliable for extended periods of time
(months) per year at the SXR beamline for experiments conducted within the experimental group of A.
Nilsson, by other SXR collaborators as well as by outside users via LCLS experiment proposals.
The UHV end station, outlined in the figure below, is designed for soft x-ray photoelectron spectroscopy
(PES), x-ray emission spectroscopy (XES) and x-ray adsorption spectroscopy (XAS) of surface and solid
state samples of up to 10mm in diameter having ultra-high vacuum compatibility. The end station is
equipped with an electron spectrometer (R3000, VG-Scienta) for PES, partial electron yield detector for
XAS, and a soft x-ray emission spectrometer housing 2 gratings for photon energies from about 220 eV to
630 eV with a maximum resolving power of about 2000. A horizontally mounted manipulator (VG
Omniax) allows transfer of the sample(s) between the preparation chamber and the analysis chamber.
Sputtering facilities, evaporation sources, mass spectrometer and LEED optics are available in the
preparation chamber. A pulsed gas delivery sustem allows an accurate deposition of gases in a reproducible
manner.
The sample setup, illustrated in the figure below, is optimized for studies of adsorbate systems prepared on
single crystal surfaces. A single crystal is mounted onto a Cu block via a pair of Ta (or W) wires. The Cu
plate is electrically isolated from the sample holder by a sapphire spacer and thus a voltage can be applied.
Temperatures between 30 and 1500 K can be achieved through cooling with liquid N2 (He) and heating
performed by electron bombardment (with or without a biaz). Temperature is measured by thermocouples
spot welded onto the sample or via Ta foil.
This endstation is currently installed and commissioned at BL 13-2 at SSRL, and fully equipped for
experiments at synchrotron radiation sources. To accommodate LCLS experiments a number minor
modifications will be installed. Of particular importance is the integration of fast cameras with the electron
spectrometer and x-ray emission spectrometer. Since both of these instruments have a similar detection
system (MCP + phoshour plate and a camera mounted on the outside of vacuum), we intend to use the
same solution for both spectrometers.
30

Dear SXR representative,
This letter describes the intention to install, operate and support two fully equipped endstations at the Soft
X-ray for Material Science (SXR) beamline at LCLS. The endstations will comply with the requirements
outlined in chapter 5 of the SXR Technical Design Report and be available for extended periods of time
(months) per year at the SXR beamline for experiments conducted within the experimental group of Z. X.
Shen and Zahid Hussain, by other SXR collaborators as well as by outside users via LCLS experiment
proposals. The capabilities of these two endstations are described in the following:
Resonant X-ray Scattering Endstation:
The photos of the actual chamber are shown in the following figure. This endstation is in an ultra-high
vacuum (UHV) environment capable of maintaining a base pressure better than 10 -9 Torr. A UHV
compatible sample loading/transfer system is installed for the solid state samples. A motorized sample
stage allows the sample to be rotated azimuthally about its surface normal. This sample stage (also shown
in the following figure) is thermally contacted to a temperature control system, which consists of a liquid
Helium cryostat and a heater, allowing the sample temperature to change from 15 K to 400 K. The sample
also has three translational degrees of freedom (through manipulator) and two rotational degrees of
freedom (through differentially pumped rotary seal).
This endstation equips with three different types of detectors, including a channeltron, a photodiode, and a
multi-channel-plate. These detectors are mounted inside the vacuum chamber on a fully motorized detector
stage, which allows those detectors to move in both horizontal (360 degrees) & vertical (45 degrees)
scattering planes. Such capability can be used to efficiently search the superlattice reflections in a wide
range of reciprocal space. X-ray absoption (XAS) can also be performed by measuring the total fluorescent
yield and total electron yield (sample-to-ground current).
This endstation has been assembled and utilized at the Advanced Light Source (ALS), Lawrence Berkeley
National Laboratory, and fully equipped for experiments at synchrotron radiation sources. To accommodate
LCLS experiments a number of modifications will be carried out. The modifications include the installation
of in-vacuum fast parallel readout fast CCD camera for recording the image on pulse by pulse basis, a
motorized six-strut system for remote chamber alignment and an additional 7 Tesla YBCO puck
underneath the sample for field-dependence studies.
31

Resonant X-ray Inelastic Scattering Endstation:
An illustration of the chamber and a photo of the actual endstation are shown in the panel (a) of the
following figure. This endstation is in an ultra-high vacuum (UHV) environment capable of maintaining a
base pressure better than 5 x 10 -10 Torr. A UHV compatible sample transfer system is designed for the solid
state samples. A motorized sample stage is designed to rotate the sample azimuthally about its surface
normal. This sample stage is thermally contacted to a temperature control system, which consists of a liquid
Helium cryostat and a heater, allowing the sample temperature to change from 15 K to 400 K.
The endstation will be equipped with a emission spectrograph with a maximum resolving power of 2200
( ~1400 for a source image of 10 μm x10 μm, the focus beam spot of SXR beamline). The emission
spectrograph will be supported by a guide rail and can be mounted on five possible mounting ports with an
angular interval of 30 degrees in the horizontal scattering plane. The rotary stage underneath the
experimental chamber can rotate both the chamber and emission spectrograph by +/- 15 degrees with
respect to the incoming x-ray beam. This design allows the users to probe a wide range of momentum
transfer (for instance, see panel (b) of the following figure which is calculated at Mn L edge). The
spectrograph is necessary for the time-resolved pump-probe inelastic scattering experiment, as it can record
a spectrum across a range of energy loss simultaneously. This endstation is also equipped with a
photodiode for measuring the total fluorescence yield. The sample stage is electrically isolated from ground
such that the total electron yield can be recorded.
This endstations has been currently assembled and tested at the Advanced Light Source (ALS), Lawrence
Berkeley National Laboratory, and fully equipped for experiments at synchrotron radiation sources. To
accommodate LCLS experiments a number of modifications will be implemented. These include a shutter
system, a motorized six-strut system for remote chamber alignment, and fast parallel readout fast CCD
camera for recording the image on pulse by pulse basis.
32

Dear SXR representative,
This letter describes the intention to install, operate and support a fully equipped endstation for photon-ion
and photon-electron coincidence measurements at the Soft X-ray (SXR) beamline at LCLS. The endstation
will comply with the requirements outlined in chapter 5 of the SXR Technical Design Report and will be
available for extended periods of time (months) per year at the SXR beamline for experiments conducted
within the CFEL experimental group, by other SXR collaborators as well as by outside users via LCLS
experiment proposals.
The endstation, outlined in the figure below, is designed to explore the interaction of intense soft x-ray
radiation with various targets of increasing complexity and size, ranging from atoms and (laser-aligned)
molecules to nano-particles such as clusters and biological targets, which can be injected by a supersonic
atomic, molecular or cluster jet or a particle injector. The endstation is equipped with two large-area,
single-photon counting pnCCD detectors, which collect scattered or fluorescent photons with an energy
resolution of 40 to 200 eV between 100 eV and > 10 keV at a frame read-out rate up to 120 Hz. A variablesized
hole in the center of the first CCD allows the direct FEL beam (and a high-power pump laser, if
applicable) to pass through the detector, while the small-angle scattering signal within the area of the hole
can be detected on the second CCD. A specially-designed ion and electron spectrometer (“reaction
microscope”) mounted perpendicular to the FEL beam direction detects ions and/or electrons created by the
interaction of the intense soft x-rays with the sample with a large solid angle, and allows measuring their
kinetic energies and emission directions. In order to account for different operation modes, the reaction
microscope is equipped with a delay-line anode for single particle detection as well as a phosphor
screen/CCD camera for applications where several hundred or thousand electrons or ions are created by
each FEL pulse.
The unique combination of large-area, single-photon counting pnCCD detector and advanced reaction
microscope thus allows, for the first time, fluorescence-photoion and/or fluorescence-photoelectron
coincidence experiments on atoms, molecules, clusters, and nano-particles. Additionally, the endstation is
fully equipped for coherent diffractive imaging experiments on biological and other targets beyond the
radiation damage limits. Coulomb explosion and damage processes can hence be studied directly via the
correlated measurement of the fragment ions as well as with respect to their influence on the diffraction
patterns. The whole set-up is constructed highly modular and flexible: the entire reaction microscope can
be easily removed, various gas-jet, droplet-jets, aerodynamic lenses as well as fixed targets or specific
types of detection devices such as a Thomson parabola spectrometer for instance can be inserted in a
straight forward manner.
The endstation is currently being designed at CFEL and will be tested at FLASH early next year prior to its
use at LCLS, which is envisioned as early as mid-2009.
33

Dear SXR representative,
this letter describes the long-term intention to install, operate and support a fully equipped endstation
dedicated to the investigation of the interaction of light with dilute clouds of highly charged ions (HCI),
trapped in an Electron Beam Ion Trap (EBIT). A significant fraction of this research is concerned with
energies up to 2 keV and requires a monochromatized photon beam such that the Soft X-ray (SXR)
beamline at LCLS is an ideal and presently the only suitable location. While experiments are anticipated to
be performed initially with an existing machine which was already successfully operated at the FLASH the
collaboration intents to apply for a dedicated LCLS-based user endstation available for a broad community.
Its potential applications cover experiments in the fields of astrophysics, in the physics of plasmas, nuclei
and to the tests of fundamental forces and symmetries. The envisioned definitive, dedicated endstation will
comply with the requirements outlined in chapter 5 of the SXR Technical Design Report and will be
available for extended periods of time (months) per year at the SXR beamline for experiments conducted
within the CFEL experimental group, by other SXR collaborators as well as by outside users via LCLS
experiment proposals.
The EBIT endstation, outlined in the figures below, is especially designed to explore the interaction of
photons with highly charged ions in any desired charge state. Scientific goals include: The determination of
photoionization and photoabsorption cross sections, both of utmost importance for astrophysics; the
production, trapping and investigation of photoionized samples of strongly correlated warm, dense plasmas,
this point representing one of the forefront topics of present plasma physics research; the direct photoexcitation
of low-lying nuclear levels, which would imply a significant advance in nuclear physics. Finally,
precision measurements on the electronic level structure of HCIs shall be performed to test fundamental
theories (QED) and parity non-conservation in the neutral currents sector. The endstation consists of a fully
operational EBIT. It is equipped with various spectrometers for visible, VUV, X-ray or fluorescence photon
detection. Eight ports transverse to the photon beam propagation direction give access for intense lasers
into the trap region as well as for injecting clusters, droplets or any other gaseous targets. One axial port
permits ion extraction and charge analysis along the trap axis. The complete endstation needs about 2.5 x
1.5 m 2 flour space and weights about 1.5 tons. It is further equipped with diagnostic tools to guarantee the
overlap between the LCLS beam with the ion cloud and to measure the photon beam energy in a pulse-bypulse
mode.
The present transportable EBIT endstation operates routinely at FLASH as well as at the BESSY
synchrotron. It is therefore ready to be transferred to the SXR beamline at any time in the year 2009.
Provided that the LCLS management and Proposal Review Committee supports the general EBIT user
endstation, the collaboration would start seeking money within 2008 in order to commission a new,
dedicated user endstation by the year 2010.
34

Left: Cross section of the EBIT with the FEL beam entering into the trap region axially, through the
electron beam dump along the magnetic field lines from the left. Fluorescence or any other radiation can be
detected through either one of the eight transverse ports as indicated. Right: Picture of the EBIT during
successful operation in a recent beamtime at FLASH.
35

Dear SXR representative,
This letter describes the intention to install, operate and support a fully equipped endstation at the Soft X-
ray for Material Science (SXR) beamline at LCLS. The endstation will comply with the requirements
outlined in chapter 5 of the SXR Technical Design Report and will be available for extended periods of
time (months) per year at the SXR beamline for experiments conducted within the experimental
collaboration between S. Techert (MPI for biophysical chemistry, Göttingen), A. Föhlisch (Institute for
Experimental Physics, Hamburg University), F. Hennies (Max-Lab, Lund University), by other SXR
collaborators, in particular A. Nilsson and K. Gaffney (Stanford University) and P. Wernet (Bessy, Berlin)
as well as by outside users via LCLS experiment proposals.
The experimental station to study chemical dynamics in the liquid phase is based on a differentially
pumped liquid jet system, developed at the MPI of biophysical chemistry. With a Rowland-type soft X-ray
spectrometer (Grace 3) X-ray emission spectroscopy is conducted using three gratings, effectively covering
the energy range between 50eV< hv < 1500 eV. Thus detailed investigations of the valence electronic
structure and the chemical state for chemically and biologically relevant molecular dynamics are accessible
both with resonant and non-resonant X-ray emission spectroscopy. In particular the local valence electronic
structure of carbon, nitrogen and oxygen, as well as transition and rare earth metals can be investigated.
Femtosecond temporal resolution to study photoinduced dynamics will be achieved in an optical-pump/Xray-probe
set-up, using the collinear optical incoupling and the tools for X-ray/optical cross-correlation
developed at Hamburg University. Additionally, X-ray induced radiation chemistry can be investigated
through femtosecond time resolved X-ray-pump/optical-probe spectroscopy.
The liquid beam set-up is shown below with schematic illustrations of prototypical chemical dynamics that
will become accessible with this facility. This station will be thoroughly tested at the Free-Electron LASer
at Hamburg FLASH and the X-ray spectrometer and associated diagnostics will be characterized at Max-
Lab Sweden. In particular we will develop the hardware and software, which we currently employ at
FLASH further and optimize for the operating conditions at LCLS.
Dissoziation
~ 1Å / 100 fs.
Isomerization
> 200 fs.
36

Dear SXR representative,
This letter describes the intention to install, operate and support a fully equipped endstation at the Soft X-
ray for Material Science (SXR) beamline at LCLS. The endstation will comply with the requirements
outlined in chapter 5 of the SXR Technical Design Report and be avaliable for extended periods of time
(months) per year at the SXR beamline for experiments conducted within the experimental group of J.
Stohr, by other SXR collaborators as well as by outside users via LCLS experiment proposals.
The endstation is in design for soft x-ray resonant single-shot imaging and small angle resonant scattering
with focus on magnetization dynamics and phase transitions. The system will run under HV conditions
(~10 -8 Torr). The endstation will be designed to accommodate the two large-area, single-photon counting
pnCCD detectors from the CFEL endstation, which collect scattered or fluorescent photons with an energy
resolution of 40 to 200 eV between 100 eV and > 10 keV at a frame read-out rate up to 120 Hz. A variablesized
hole in the center of the first CCD allows the direct FEL beam (and a high-power pump laser, if
applicable) to pass through the detector, while the small-angle scattering signal within the area of the hole
can be detected on the second CCD. Optional an in-vacuum CCD with center beam stop for low fluence
(monochromatic) experiments will be integrated.
Compatible to the design of the existing soft x-ray coherent imaging endstation at SSRL beamline 13-3, see
figure below, the main chamber will have an Omniax manipulator with cryostat providing a temperature
range of 30-400K. Removable pole pieces of an ex-situ electromagnet will provide magnetic fields (~.1
Tesla) applied along the x-ray beam path. Sample transfer will be possible without breaking vacuum
through a load lock chamber containing several sample holders.
The differential pump section between the beamline and endstation will be equipped with beam diagnostics,
magnetic filter (for circular polarized x-rays) and fast beam shutter.
Figure: Existing endstation at SSRL beamline 13-3 for Coherent X-ray Imaging.
37